Electric Dipoles in Liquid Dielectrics.

J. Herweg. Die elektrischen Dipole in flüssigen Dielektrizis. Phys. ZS. 21, p. 572 (1920).

The qualitative theory of dielectrics proposed in 1912 by Debye explains deviations from the classical Clausius-Mossotti formula observed in certain liquid dielectrics. According to Debye’s theory, within molecules, in addition to elastically bound electrons, there are also rigid electric dipoles with a constant moment. The experimental work of Ratnowski (in 1913) seemed to confirm the existence of such dipoles in certain organic compounds; but, as the author of the note under review believes, an error had crept into Ratnowski’s calculations, and, after its correction, the results of the work cannot be recognized as agreeing with Debye’s theory. Undertaking a new experimental verification of Debye’s theory, the author of the note first of all comes to the conclusion that: 1) one cannot consider the polarization of a liquid dielectric while neglecting the displacement of elastically bound electrons; 2) one cannot expect a definite answer from experiments if they are carried out with the degree of accuracy attained by Ratnowski.

The latter consideration led the author of the note to develop a very refined method for observing very small changes in the dielectric constant. The idea of this method is as follows. The plates of a small capacitor, connected into a circuit in which electric oscillations are excited by means of a generator tube, are immersed in the liquid dielectric under investigation.

This circuit is coupled (weakly) with another similar circuit, in which (by means of a variable capacitance) a number of oscillations is established that differs from the first circuit by 1000 oscillations per second. A telephone receiver, included in the tube circuit of the second circuit, makes it possible to follow the number of beats by comparing the sound of the telephone with the sound of a tuning fork making 1000 oscillations per second. This makes it possible to observe changes in the number of oscillations very accurately. Thus, for example, if the first circuit is tuned to 1,000,000 oscillations per second, and the second to 1,001,000, then the sound of the telephone coincides with the sound of the tuning fork; if, however, the number of oscillations of the first circuit changes by only \(10^{-6}\), i.e. becomes equal to 1,000,001, — between the sound of the telephone and

... in the chamber, beats will arise (1 beat per second). Starting from W. Thomson’s formula, it is not difficult to calculate that a change in the capacitance of the condenser in the first circuit by \(10^{-5}\) will produce 5 beats per second.

The rotation of dipoles, as in Ratnowsky’s experiments, was produced by means of an electrostatic field between the plates of the condenser. The field strength could reach \(100\ C.G.S.\)

The author attempted to detect dipoles in mixtures of 10%–20% amyl alcohol with benzene, with which Ratnowsky worked, but obtained no positive result. Meanwhile, in ethyl ether the change in the dielectric constant with a change in field strength proved to be of the same order as that required by Debye’s theory, taking for the moment of the electric dipole in the molecule of ethyl ether the value

\[ m = 11.8 \cdot 10^{-19}. \]

When the field strength was increased from zero to \(95.2\ C.G.S.\), the dielectric constant decreased by

\[ \Delta \varepsilon = 6.7 \cdot 10^{-6}. \]

A theoretical calculation carried out by the author gave the value

\[ \Delta \varepsilon = 6 \cdot 10^{-6}. \]

The difference between the theoretical and experimental values does not exceed the limits of error.

Thus, the author of the paper reviewed has succeeded in showing that the dielectric constant of certain liquids changes with the magnitude of the electrostatic-field strength, and this change can be satisfactorily explained by the rotation of molecular electric dipoles, whose potential energy thereby decreases, as required by Debye’s theory.

Vas. Shuleikin.